HIV and hepatitis B (HBV) co-infection is common and significantly increases the progression to liver disease and death. Populations at risk for HIV/HBV co-infection include those in Africa and Asia, regions where HIV continues to spread in the setting of HBV hyperendemicity. Lamivudine (3TC), an antiretroviral (ARV) effective against HBV, is included in 4 WHO first-line ARV regimens, yet there is little data on HIV/HBV treatment outcomes in these areas. Evidence suggests that HBV genotype is predictive of response to therapy. Likewise, HBV and factors such as HBV viral load, genotype, and resistance may contribute to hepatotoxicity during ARV therapy in Africa, and yet little is known about the role of HBV in HIV treatment outcomes. Consistent with the NIAID's goal to foster international collaborative research on HIV, its coinfections, and therapeutic strategies in resource limited settings, the aims of this study are 1) to characterize the prevalence of HBV infection and genotypes in HIV infection 2) to examine the role of HBV genotype in HBV suppression and resistance formation and 3) to examine the association of HBV infection to severe hepatotoxicity in HIV individuals on ARVs in South Africa. A primary aim is to support the applicant's patient-oriented research career development through mentoring, training in molecular epidemiology, and the receipt of a Master's Degree in Public Health. The proposed research is a prospective cohort study nested within the Cape Town AIDS Cohort, an NIH funded study of HIV treatment in South Africa. This population will be characterized for HIV/HBV coinfection and HBV genotype. HBV suppression and resistance will then be determined at 1 year in 135 HIV/HBV coinfected individuals receiving 3TC containing ARV therapy. In an estimated 135 with hepatotoxicity, the presence of HBV infection and virologic factors will be compared to 1215 controls. Potential confounders such as hepatitis C, liver disease, ARV regimen, and other causes of hepatotoxicity will be included in the statistical models of HBV effect on treatment outcomes. The long-term goals of this research are to identify predictors of therapeutic outcomes in HIV/HBV coinfection. This study is the first to examine the role of HBV genotype in HIV/HBV treatment outcomes in a resource limited setting. If HBV virologic factors are associated with treatment outcomes in HIV/HBV coinfection, then identifying such factors may improve future diagnostic and therapeutic strategies in HIV/HBV co-infection.